A charging and discharging control system for the power battery of a new energy sightseeing vehicle

Abstract

This invention discloses a charging and discharging control system for a new energy sightseeing vehicle's power battery, relating to the field of battery charging and discharging technology. It includes a power generation prediction module, a demand analysis module, a decision management module, and a dynamic monitoring module. The power generation prediction module extracts future travel and weather data from the sightseeing vehicle management system and a meteorological platform. Its key technical points are: through a standardized API interface, it collaborates with the existing sightseeing vehicle management system to collect travel data such as departure time, route, load, and driving style; it combines this with meteorological data to segment the travel and correlate weather data; after power generation prediction, demand analysis, and decision management, it generates a charging and discharging control strategy and feeds it back to the sightseeing vehicle management system; it utilizes an edge network to monitor execution data in real time to form a closed loop, achieving dynamic matching of energy flow and operational scheduling, improving energy utilization efficiency and operational economy, demonstrating good performance and promising application prospects.

Description

Technical Field

[0001] This invention relates to the field of battery charging and discharging technology, specifically to a charging and discharging control system for a power battery of a new energy sightseeing vehicle. Background Technology

[0002] With the continuous advancement of solar photovoltaic power generation technology, the conversion efficiency of solar panels will continue to improve, and the cost will be further reduced, making the application of photovoltaic power generation in new energy tourist sightseeing vehicles increasingly common.

[0003] New energy sightseeing vehicles, as a type of clean energy transportation, are characterized by high energy efficiency and environmental friendliness, and are becoming the main development direction for future transportation.

[0004] New energy sightseeing vehicles in scenic areas have become the preferred mode of transportation due to their green and environmentally friendly characteristics. Their efficient operation relies heavily on the protection of the charging and discharging control system. This system monitors the battery status in real time, precisely regulates the charging current and voltage, effectively ensuring battery safety and extending its service life. At the same time, it improves energy utilization and vehicle range by optimizing the discharging strategy.

[0005] During the off-season, the number of tourists in scenic areas drops sharply, leading to a significant decrease in the frequency of use of new energy sightseeing vehicles, with many vehicles remaining idle. Meanwhile, the photovoltaic power generation systems within the scenic areas continue to operate. If the generated electricity cannot be consumed in a timely manner, it could be wasted. In this situation, the electricity generated by the new energy sightseeing vehicles can be sold through vehicle-to-grid (V2G) technology.

[0006] When the scenic area's own electricity demand is low, these sightseeing vehicles can supply the stored electrical energy back to the grid. During this process, the charging and discharging control system plays a continuous role, precisely adjusting charging and discharging parameters by monitoring the battery status in real time to ensure safe, stable, and efficient bidirectional energy flow. Therefore, people have invented some electric vehicle charging and discharging control systems.

[0007] An existing patent application, CN110171323A, entitled "A V2G-based Electric Vehicle Charging and Discharging Control System and Usage Method," describes an invention comprising: a vehicle controller, an electric drive control system, a power battery, and a battery management system. The electric drive system includes a DC / AC converter and a motor. The vehicle controller receives external information unidirectionally and exchanges information bidirectionally with both the battery management system and the DC / AC converter. The battery management system includes a power battery electrically connected to the DC / AC converter. The DC / AC converter is electrically connected to the motor, and the motor is electrically connected to the power grid. When the motor is disconnected from the power grid, the DC / AC converter controls the motor's speed and torque output. When the motor is connected to the power grid, the vehicle controller integrates information and issues charging and discharging commands, at which point the DC / AC converter charges and discharges the power battery. This invention has a simple structure, effectively improves the utilization of in-vehicle space and charging / discharging power, and protects user interests.

[0008] The central idea of ​​the solution described in the aforementioned patent is to realize the charging and discharging of electric vehicles, and to further study the charging and discharging of electric vehicles to achieve the adjustment of the charging and discharging speed. However, it is only a simple adjustment of the charging and discharging process, and it does not consider the overall situation. It does not take into account whether the sightseeing vehicle is in use, external electricity price fluctuations, etc. Therefore, its effect is relatively simple. It is equivalent to an independent system with little connection with other external systems. It cannot be connected to the existing sightseeing vehicle management system to achieve overall scheduling. Therefore, its applicability is relatively low, it cannot be comprehensively analyzed, and it does not meet the requirements of new energy sightseeing vehicles. To this end, we have developed a new energy sightseeing vehicle power battery charging and discharging control system. Summary of the Invention

[0009] (a) Technical problems to be solved

[0010] To address the shortcomings of existing technologies, this invention provides a charging and discharging control system for the power battery of a new energy tourist sightseeing vehicle. This system collaborates with existing sightseeing vehicle management systems via a standardized API interface, collecting travel data such as departure time, route, load, and driving style. It combines this data with meteorological data to segment the journey and correlate weather data. After power generation forecasting, demand analysis, and decision management, a charging and discharging control strategy is generated and fed back to the sightseeing vehicle management system. An edge network is used to monitor the execution data in real time to form a closed loop, achieving dynamic matching of energy flow and operational scheduling. This improves energy utilization efficiency and operational economy, demonstrating good performance and promising application prospects, and effectively solving the problems mentioned in the background technology.

[0011] (II) Technical Solution

[0012] To achieve the above objectives, the present invention provides the following technical solution:

[0013] A charging and discharging control system for a power battery of a new energy sightseeing vehicle includes a power generation prediction module, a demand analysis module, a decision management module, and a dynamic monitoring module.

[0014] Power generation prediction module: Extracts future travel and weather data from the sightseeing vehicle management system and meteorological platform, segments the future travel, and uses data fusion technology to mark the future weather data on the future sightseeing vehicle travel segments. It then uses a photovoltaic power generation prediction model to predict the power generation of all future sightseeing vehicle travel segments.

[0015] Demand Analysis Module: Construct a random forest regression model, process future sightseeing bus trip data and future weather data, and input them into the model to predict the energy consumption demand of the sightseeing bus in the future trip.

[0016] Decision Management Module: Acquires power generation, energy consumption demand, and current power consumption data of sightseeing vehicles, calculates comprehensive control values, compares comprehensive control values ​​with corresponding control intervals, and formulates control strategies based on the comparison results;

[0017] Dynamic monitoring module: Executes control strategies, acquires control strategy execution data in real time through edge network, generates and applies optimization strategies through dynamic monitoring model, and dynamically manages the charging and discharging of sightseeing vehicles.

[0018] Furthermore, standardized application programming interface (API) technology is used to collect future trip and future weather data. Future trip data includes departure time, route data, estimated passenger load, and driver style data, while future weather data includes temperature, humidity, wind speed, and weather conditions.

[0019] Furthermore, the future itinerary is segmented, including segmenting the stationary state of the sightseeing vehicle using time series analysis and segmenting the moving state of the sightseeing vehicle using spatial clustering algorithms, to obtain the sightseeing vehicle itinerary sub-segments.

[0020] The steps for using data fusion technology to label future weather data with segments of the future sightseeing bus itinerary are as follows:

[0021] The data of the future sightseeing bus trip segments are deeply structured so that the future sightseeing bus trip segments contain unique identifiers, start and end timestamps, spatial coordinate sequences, status types and associated road segment IDs;

[0022] Standardize and convert future weather data according to a unified time granularity;

[0023] Establish a correlation between future sightseeing vehicle trip segments' timestamps and spatial coordinates and future weather data;

[0024] Spatial and temporal data are processed based on spatial and temporal interpolation.

[0025] The DTW algorithm is used to align the time sequence of travel and weather data that are misaligned due to fluctuations in vehicle speed.

[0026] A semantic network is constructed, employing named entity recognition technology from natural language processing and association rule algorithms from data mining to achieve automatic extraction and annotation of semantic tags, forming semantic tags for each segment of the future sightseeing bus itinerary;

[0027] Semantic tags are divided into basic tags and context tags.

[0028] Furthermore, the steps for predicting the power generation of the sightseeing vehicle's journey segment using a photovoltaic power generation prediction model are as follows:

[0029] Obtain semantic tags for future sightseeing vehicle itinerary segments;

[0030] Based on the basic tags of the future sightseeing vehicle itinerary segments, data in the database is retrieved, and data with the same basic tags are retained to obtain a preliminary filtered dataset.

[0031] The environmental labels of the future sightseeing bus itinerary segment are compared with the environmental labels in the initially screened dataset, and the similarity is calculated using the following formula:

[0032]

[0033] In the formula: Simx represents the similarity score. It is an adjustable parameter. , where m is the number of dimensions for the environment tags. For the weight data of the i-th environmental label dimension, Let i be the value of the i-th environment label dimension in the environment label of the future itinerary segment. To initially filter the values ​​of the i-th environment label dimension in the historical itinerary sub-segments of the dataset, Let be the intermediate similarity, exp be an exponential function with the natural constant e as the base, and b be an adjustable parameter. .

[0034] Arrange the data in descending order of similarity, select the top K groups of similarity corresponding to the power generation, and then weight the power generation to obtain the power generation for each segment of the sightseeing bus trip. The weighting formula is as follows:

[0035]

[0036] In the formula, The predicted power generation for the sightseeing bus route segment. Let be the power generation of the j-th group, e be the natural constant, and c be the attenuation coefficient, where c > 0. The objective function for c is fitted using the least squares method. For the i-th The interval between the corresponding time point and the current time.

[0037] Furthermore, the construction steps of the random forest regression model are as follows:

[0038] Acquire historical data, clean the collected data, and then encode the categorical data;

[0039] The historical data was processed using the min-max normalization method, and the processed data was divided into training and test sets.

[0040] From the training set, a training set for each decision tree is constructed by random sampling with replacement.

[0041] Randomly select a subset of features from all features to construct nodes for the decision tree, find the optimal split point according to the criterion of minimizing mean squared error, construct the decision tree, and form a random forest regression model;

[0042] The random forest regression model is trained using a test set, and the mean squared error is used as the evaluation metric to output a random forest regression model whose test accuracy is greater than or equal to a preset threshold.

[0043] Furthermore, the comprehensive control value = (rated power of sightseeing vehicle + power generation of all future sightseeing vehicle trip segments - energy consumption demand - safe power) / rated power of sightseeing vehicle × 100%, and the control range includes the charging range (-100%, BZ1) and the fluctuation range. and discharge range (BZ2, 100%).

[0044] Furthermore, the comprehensive control value is compared with the corresponding control range, and control strategies are formulated based on the comparison results, including:

[0045] If the comprehensive control value is within the charging range, that is, the first charge adopts the nighttime off-peak charging, the sightseeing vehicle is charged to full charge, and the second charge adopts dynamic charging, then the control strategy is the first charge + dynamic charging.

[0046] If the comprehensive control value is not within the charging range, then the secondary control value is calculated. The secondary control value = (current power of the sightseeing vehicle + power generation of all future sightseeing vehicle trip segments - energy consumption demand - safe power) / rated power of the sightseeing vehicle × 100%;

[0047] If the secondary control value is within the charging range, nighttime off-peak charging is adopted, and the charging amount = |secondary control value × sightseeing vehicle rated power|. At this time, the control strategy is single nighttime charging.

[0048] If the secondary control value is within the fluctuation range, the control strategy is not to perform charging and discharging treatment;

[0049] If the secondary control value is within the discharge range, then discharge is performed. The discharge amount = secondary control value × sightseeing vehicle rated power, and dynamic discharge is executed. At this time, the control strategy is dynamic discharge.

[0050] Furthermore, the dynamic charging and discharging scheme is optimized through a charging and discharging strategy model based on a deep Q-network.

[0051] The input to the charging and discharging strategy optimization model based on deep Q network during dynamic charging is the charging amount, dynamic electricity price data, and loss data corresponding to the charging speed. The output is the optimal charging power, charging time arrangement, and charging amount at each time.

[0052] The input to the dynamic discharge strategy optimization model based on deep Q-network is the discharge capacity, dynamic electricity price data, and loss data corresponding to the discharge rate. The output is the optimal discharge power, discharge time arrangement, and discharge amount at each time.

[0053] Furthermore, by using edge networks to acquire real-time data on the execution of control strategies, sensors deployed on sightseeing vehicles and charging facilities utilize edge networks to collect data on charging and discharging temperatures, current electricity prices, charging power, and discharging power during the execution of charging and discharging strategies.

[0054] When data is collected and controlled through the edge network, it is transmitted to the edge server in real time. The edge server performs preliminary cleaning and preprocessing on the data, removes outliers and duplicate data, and stores the processed data in the local cache.

[0055] Furthermore, the dynamic regulatory model includes:

[0056] Data input layer: Receives charging and discharging strategy execution data transmitted from the edge network and performs normalization processing;

[0057] Feature extraction layer: Utilizes a spatiotemporal graph neural network to extract spatiotemporal features from the data, and analyzes the correlation between charging and discharging states, electricity prices, and temperature in time series and spatial distribution;

[0058] Decision Analysis Layer: Based on a deep Q-network decision mechanism, with the goal of minimizing charging and discharging costs and maximizing energy utilization efficiency, it combines the results of the feature extraction layer to calculate the Q value under different states, evaluate the merits of the current charging and discharging strategy, and adjust the strategy based on the dual factors of electricity price and charging temperature.

[0059] Strategy output layer: Outputs optimization strategies, including discharge optimization strategies or charging optimization strategies;

[0060] Charging optimization strategies include charging power and charging time;

[0061] The discharge optimization strategy includes discharge power and discharge time.

[0062] (III) Beneficial Effects

[0063] This invention provides a charging and discharging control system for the power battery of a new energy sightseeing vehicle, which has the following beneficial effects:

[0064] 1. This invention describes a charging and discharging control system for a power battery of a new energy sightseeing vehicle. It works in conjunction with the existing sightseeing vehicle management system to collect travel data such as departure time, route, load, and driving style. It combines meteorological data to segment the journey and correlate weather data. After power generation prediction, demand analysis, and decision management, it generates a charging and discharging control strategy and feeds it back to the sightseeing vehicle management system. It uses an edge network to monitor the execution data in real time to form a closed loop, realizes dynamic matching of energy flow and operation scheduling, improves energy utilization efficiency and operational economy, has wide adaptability, good use effect, and has good application prospects.

[0065] 2. This invention describes a charging and discharging control system for the power battery of a new energy sightseeing vehicle. During use, it improves energy management accuracy through multi-dimensional data integration and spatiotemporal modeling: it collects sightseeing vehicle trip data and meteorological data, uses time series analysis and spatial clustering algorithms to segment the trip into stationary / moving states, achieves deep coupling between trip segments and weather data through data fusion technology, and combines similarity retrieval and exponential weighting algorithms from a photovoltaic power generation prediction model to provide more granular spatiotemporal power generation prediction support with stronger environmental adaptability for charging and discharging decisions. This significantly improves the scientific nature and reliability of energy flow management for new energy sightseeing vehicles, resulting in good overall performance.

[0066] 3. This invention describes a charging and discharging control system for the power battery of a new energy sightseeing vehicle. Through multi-source data fusion and a photovoltaic power generation prediction model, it achieves accurate prediction of power generation in spatiotemporal scenarios. Furthermore, by training cleaned and encoded historical data using a random forest regression model, it improves the accuracy of energy demand prediction. Based on comprehensive control values ​​and multi-interval thresholds combined with strategies such as off-peak electricity at night, dynamic charging and discharging, and V2G discharging, it achieves power balance and cost optimization. Further, by collecting data in real time through an edge network and generating optimization strategies through a spatiotemporal graph neural network and a deep Q-network, it achieves multi-dimensional dynamic optimization of the charging and discharging of the new energy sightseeing vehicle, significantly improving energy utilization efficiency, battery life, and grid coordination capabilities. It demonstrates good performance and promising application prospects. Attached Figure Description

[0067] Figure 1 This is a flowchart illustrating the charging and discharging control system for a new energy tourist sightseeing vehicle according to the present invention. Detailed Implementation

[0068] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0069] The core idea of ​​the solutions described in the existing patents is to determine the state of the electric vehicle's power battery, to determine the charging and discharging state of the power battery based on the corresponding mechanism, and to further study and improve the charging and discharging of electric vehicles to achieve the adjustment of the charging and discharging speed.

[0070] However, in practical use, this solution has the following limitations:

[0071] Limited applicability: This method is suitable for charging and discharging in daily scenarios (going out early and returning late to avoid peak and off-peak electricity usage), but it cannot achieve timed discharge during peak electricity usage or charging only during off-peak electricity usage for long-term use in scenic areas where daily operation varies.

[0072] Low compatibility: This type of system mainly manages the power battery independently and cannot be dynamically planned and adjusted in accordance with usage requirements and electricity prices.

[0073] This invention overcomes the limitations of existing technologies, such as data singularity, extensive prediction, and static control, through multiple innovations in multi-source data fusion, photovoltaic power generation prediction, energy demand prediction, and dynamic decision-making and optimization. It constructs a fully intelligent system encompassing data acquisition, spatiotemporal modeling, intelligent decision-making, and dynamic optimization, significantly improving the precision and efficiency of energy management for new energy sightseeing vehicles. It also provides reusable technical specifications for the energy charging and discharging management of new energy sightseeing vehicles, laying the technical foundation for charging and discharging management.

[0074] Example 1

[0075] Please see Figure 1 This embodiment provides a charging and discharging control system for the power battery of a new energy tourist sightseeing vehicle. The specific solution is as follows:

[0076] 1. System architecture;

[0077] 1.1 Data Acquisition;

[0078] The data acquisition system is responsible for collecting various types of data related to the charging and discharging management of sightseeing vehicles in real time, and deploying multiple sensors on sightseeing vehicles and charging facilities.

[0079] Data acquisition mainly collects vehicle status, environmental data, and charging / discharging parameters. Vehicle status relies on vehicle status sensors, which collect battery status data such as remaining battery power, battery voltage, and current. These sensors monitor battery health and charging / discharging capabilities in real time, providing a basis for developing charging / discharging strategies. For example, by obtaining the remaining battery power, the current energy reserve of the vehicle can be determined.

[0080] Environmental data relies on environmental sensors, including temperature sensors and light intensity sensors, which are used to collect environmental data such as charging temperature and solar radiation intensity. When using these sensors, their placement depends on the characteristics of the scenic area. For example, the influence of wind needs to be considered in mountainous and plain areas, so additional corresponding sensors need to be deployed.

[0081] For scenic areas with lower requirements, environmental data can be obtained directly from the meteorological platform. However, this method yields relatively little data and has lower accuracy. This method is not suitable for scenic areas with high requirements.

[0082] Charging and discharging parameters are acquired through sensors installed on the charging equipment. These sensors can monitor parameters such as charging power, discharging power, and actual charging and discharging amount in real time, so as to understand the real-time status of the charging and discharging process and provide data support for subsequent strategy optimization.

[0083] 1.2 Transmission Network;

[0084] The network layer bears the heavy responsibility of data transmission, employing edge network technology to achieve efficient and stable transmission of data from the sensing layer. It primarily uses communication technologies such as 5G networks and low-power wide-area networks (LPWANs) to rapidly transmit data collected by sensors to edge servers. 5G networks, with their high speed and low latency, ensure timely transmission of data with high real-time requirements (such as urgent battery anomaly data); LPWANs are suitable for data transmission from power-sensitive sensors with smaller data volumes, reducing overall system energy consumption. Simultaneously, the edge network possesses certain local computing and data processing capabilities, enabling preliminary data cleaning and preprocessing during transmission, reducing invalid data transmission and improving data transmission efficiency and quality.

[0085] Data collected from the sightseeing vehicle management system and the meteorological platform is processed using standardized application programming interfaces (APIs). Communication connections are established via RESTful or SOAP protocols based on the system's Web API specifications. This interface technology can accurately locate and retrieve future travel schedule data stored in the sightseeing vehicle management system, including structured data such as vehicle operation plans, routes, departure times, and stop information. Similarly, data acquisition from the meteorological platform relies on its open data interface. Future weather data is requested according to specified data formats (such as JSON and XML), covering multi-dimensional meteorological information such as temperature, humidity, light intensity, wind speed and direction, and precipitation probability. This data interface technology ensures efficient and stable data source integration, laying the foundation for subsequent data collection.

[0086] When collecting data, ETL technology is used to perform in-depth processing on the connected data.

[0087] For example, in the data extraction stage, for the sightseeing vehicle management system, future travel data is extracted periodically or in real-time according to preset time periods and data filtering rules; for meteorological platform data, precise extraction is performed according to geographical regions and time ranges. In the data conversion stage, the extracted heterogeneous data is formatted uniformly, such as standardizing the time format of different data sources and unifying the units of numerical data. At the same time, the data is cleaned to remove duplicate records, correct erroneous data, and fill in missing values. Finally, through data loading operations, the processed future travel data and weather data are stored in the system's data warehouse or a designated database table, providing standardized and high-quality data support for subsequent data analysis and model training.

[0088] For the data on the execution of control strategies, sensor nodes and communication modules are deployed on each sightseeing vehicle using Internet of Things (IoT) technology. Onboard sensors collect real-time data on the vehicle's operating status, such as current location, speed, and battery level. This data is transmitted in real-time to the sightseeing vehicle management system via wireless networks (such as 4G / 5G and NB-IoT), ensuring that future travel data in the system is updated in real time. Simultaneously, the meteorological platform relies on a widely distributed network of meteorological monitoring stations and IoT sensor networks, such as meteorological satellites, ground weather stations, and weather radars, to collect meteorological data in real time. This data is then aggregated and processed by a data center and ultimately provided to the system through a data interface, ensuring the timeliness and accuracy of future weather data.

[0089] 1.3 Data Analysis

[0090] Data analysis is the core of this system, integrating multiple key models and data processing functions to achieve in-depth data analysis and strategy optimization.

[0091] Data analytics includes data preprocessing, namely data cleaning, integration and storage, as well as standardization.

[0092] Data analysis also includes various models, such as power generation prediction, random forest regression model operation, decision making in the decision management module, and dynamic management in the dynamic monitoring model.

[0093] 1.4 Strategy Application

[0094] The strategy application mainly involves applying the data analysis results to real-world scenarios and enabling the monitoring and management of the charging and discharging process of sightseeing vehicles.

[0095] For example, receiving optimized charging and discharging strategies can control the operating parameters of the sightseeing vehicle's charging and discharging equipment, such as adjusting the charging power, starting or stopping the charging and discharging operation, to ensure that the sightseeing vehicle charges and discharges according to the optimal strategy and achieves rational use of energy.

[0096] For example, by using an edge server to monitor various data during the charging and discharging process of the sightseeing vehicle in real time, an alarm will be issued immediately if any abnormal data is detected (such as excessively high charging temperature or excessive fluctuations in charging and discharging power). The abnormal information will be uploaded to the management platform, and the management personnel can take timely measures based on the alarm information to ensure the safe operation of the vehicle and equipment.

[0097] For example, it provides managers with a visual interface to display information such as the sightseeing vehicle's operating status, charging and discharging data, and strategy execution effects. At the same time, the management platform regularly evaluates and optimizes the dynamic monitoring model based on real-time monitoring data and user feedback, adjusting model parameters and algorithms to achieve continuous improvement and optimization of the system.

[0098] 2. System Solution

[0099] A charging and discharging control system for the power battery of a new energy sightseeing vehicle, such as Figure 1 As shown, it includes a power generation forecasting module, a demand analysis module, a decision management module, and a dynamic monitoring module:

[0100] This system is based on data from the existing sightseeing vehicle management system. Only when the sightseeing vehicle management system formulates the corresponding sightseeing vehicle plan can subsequent analysis and judgment be realized. Therefore, the foundation of this system is data collection and processing.

[0101] In order to ensure operation, sufficient data must be collected in this system. The data collection and processing are based on the power generation prediction module. The power generation prediction module extracts future travel and future weather data from the sightseeing vehicle management system and the meteorological platform, segments the future travel, and uses data fusion technology to mark the future weather data on the future sightseeing vehicle travel segments. The power generation prediction model predicts the power generation of all future sightseeing vehicle travel segments.

[0102] Standardized application programming interface (API) technology is used to collect future trip and weather data. Future trip data includes departure time, route data, estimated passenger load and driver style data, while future weather data includes temperature, humidity, wind speed and weather conditions.

[0103] To ensure the accuracy and periodicity of the data, a period of 1 day will be adopted in the future.

[0104] Therefore, the collected future itinerary data corresponds to the sightseeing bus for the next day, generally including departure time, route, stops, estimated travel time, and number of passengers. In order to ensure the operation of the scenic area, traffic control is usually adopted so that the sightseeing bus can operate according to the settings under normal circumstances, thus ensuring the accuracy of the system analysis.

[0105] For example, vehicle number 003 departs at 9:00 AM, with a total of 4 stops, a 2-hour journey, a 3-minute stop at the 2nd stop, a 5-minute stop at the 3rd stop, a 20-minute stop at the last stop before returning, and a total distance of 50 kilometers. Similarly, vehicle number 003 departs at 2:00 PM, with a total of 4 stops, a 2-hour journey, a 3-minute stop at the 2nd stop, a 5-minute stop at the 3rd stop, a 20-minute stop at the last stop before returning, and a total distance of 50 kilometers.

[0106] Future weather data can be obtained through a meteorological API, including future weather forecasts along the travel route, such as date, time, location, weather conditions (sunny / rainy / cloudy, etc.), and solar radiation intensity (W / m²). 2 (e.g., temperature, humidity, wind speed).

[0107] For example, at 9:00 AM, the weather is sunny with a solar radiation intensity of 800 and a temperature of 25°C; at 2:00 PM, the weather is sunny with a solar radiation intensity of 1000 and a temperature of 33°C.

[0108] Once the relevant data is collected, since the vehicles are not in a uniform state, it is necessary to further subdivide the vehicle operation, that is, to divide the formation of the day into multiple stages, and then evaluate and analyze each stage separately.

[0109] The process of segmenting the future itinerary includes segmenting the stationary state of the sightseeing vehicle using time series analysis and segmenting the moving state of the sightseeing vehicle using spatial clustering algorithms, thus obtaining the sightseeing vehicle itinerary sub-segments.

[0110] The steps for using data fusion technology to label future weather data with segments of the future sightseeing bus itinerary are as follows:

[0111] The data of the future sightseeing bus trip segments are deeply structured so that each segment contains a unique identifier, start and end timestamps, spatial coordinate sequence, status type and associated road segment ID, which facilitates subsequent segmentation processing.

[0112] After data consistency processing, time series analysis is performed on the collected vehicle speed data. A speed threshold (e.g., 0.5 km / h) is set. When the vehicle speed is below this threshold and the stationary time exceeds 2 minutes, it is determined to be in a stationary state. Using time series analysis algorithms, features are extracted from the time series data of the stationary state, and statistical measures such as the mean, variance, and autocorrelation coefficient are calculated to analyze the stability and periodicity of the stationary state. When the characteristics of the time series show significant changes, segment nodes of the stationary state are identified. The continuous stationary state is divided into different sub-segments, and the start timestamp, end timestamp, duration, and vehicle position coordinates of each sub-segment are recorded.

[0113] The geographic coordinate sequence during the vehicle's journey is obtained based on GPS positioning data. Spatial clustering algorithms such as Density Peak Clustering (DBSCAN) are used to perform cluster analysis on the geographic coordinate points. Based on a pre-set appropriate neighborhood radius and density threshold, adjacent geographic coordinate points with high density are grouped into one class, forming different spatial clusters. Each cluster corresponds to a specific area on the sightseeing vehicle's journey path, such as a road within a scenic area or a section of road around a certain attraction. The process of the vehicle moving from one cluster to another is regarded as a segment node of the driving state. The continuous driving process is divided into multiple sub-segments based on spatial regions. Each sub-segment contains information such as the starting cluster, ending cluster, driving path, and driving time.

[0114] Future weather data is standardized and converted according to a uniform time granularity (e.g., 10 minutes, adjusted according to accuracy) to ensure data format consistency;

[0115] The predefined future sightseeing vehicle itinerary segments are subjected to deep structuring processing. Each segment is assigned a unique identifier (UUID) and its start and end timestamps, spatial coordinate sequence (containing multiple geographic coordinate points), status type (stationary or moving), and associated road segment ID (matched via the scenic area's electronic map) are recorded. The structured itinerary segment data is then stored in a relational database for easy subsequent querying and processing.

[0116] The acquired future weather data is standardized and transformed according to a uniform time granularity, missing data is supplemented and outliers are removed. For data with discontinuous timestamps, linear interpolation is used to complete it; for abnormal meteorological data (such as temperature values ​​outside the normal range), it is corrected or removed by comparing and analyzing data from surrounding time points. At the same time, a geographic coordinate field is added to each meteorological data record, and the location of meteorological monitoring stations is aligned with the coordinate system of the scenic area map through geocoding to form a standardized meteorological dataset.

[0117] Establish a correlation between future sightseeing vehicle trip segments' timestamps and spatial coordinates and future weather data;

[0118] Based on the timestamps and spatial coordinates of future sightseeing vehicle journey segments, a correlation is established with future weather data. For stationary segments, the coordinates of their stopping positions are used as a reference to match the data of the nearest meteorological monitoring station for the same period. For moving segments, meteorological data of the corresponding time are associated with key nodes on their travel path (such as the start, midpoint, and end of the road segment). In the process of association, both time and space matching are considered to ensure the accuracy and consistency of the data.

[0119] Spatial and temporal data are processed based on spatial and temporal interpolation.

[0120] Under normal circumstances, spatial dimension interpolation uses a combination of inverse distance weighted interpolation and Kriging interpolation. However, when there are complex terrains or areas with large fluctuations in meteorological data, the Kriging algorithm is switched to improve the interpolation accuracy by modeling spatial autocorrelation through the variogram function. Since this process is frequently used in existing technologies, it will not be described in detail.

[0121] Under normal circumstances, when performing time-dimension interpolation, for travel segments spanning multiple time periods, if the time interval of the meteorological data is inconsistent with the time interval of the travel segment, the cubic spline interpolation method is used. Based on the data from the preceding and following times, a smooth curve is fitted, and the estimated values ​​of meteorological parameters at each time point within the segment are calculated to ensure the continuity of data in the time dimension.

[0122] The DTW algorithm is used to align the time sequence of travel and weather data that are misaligned due to fluctuations in vehicle speed.

[0123] To address the time misalignment between travel and meteorological data caused by fluctuations in vehicle speed, a dynamic time warping algorithm is employed for time series alignment. Using the time series of travel segments as a benchmark, the Euclidean distance matrix between these segments and the meteorological data time series is calculated. Dynamic programming is used to find the optimal curved path, elastically stretching or compressing the time axis of the meteorological data to ensure that each travel segment accurately matches the meteorological conditions of its corresponding time period.

[0124] A semantic network is constructed, employing named entity recognition technology from natural language processing and association rule algorithms from data mining to achieve automatic extraction and annotation of semantic tags, forming semantic tags for each segment of the future sightseeing bus itinerary;

[0125] Semantic tags are divided into basic tags and context tags.

[0126] The generated semantic tags are divided into two main categories: basic tags and environmental tags. Basic tags mainly include road segment ID, start and end timestamps, trip distance, driving direction, and status type, which are used to describe the basic attributes of the trip segments. Environmental tags mainly include weather conditions, solar radiation intensity, temperature, humidity, and wind speed, which are used to describe the environmental conditions corresponding to the trip segments. By classifying the semantic tags, it is easier to perform subsequent data retrieval and analysis based on the tags, which can reduce the workload of subsequent processing and improve the efficiency of analysis.

[0127] The steps for predicting the power generation of a sightseeing bus trip segment using a photovoltaic power generation prediction model are as follows:

[0128] To obtain semantic tags for future sightseeing vehicle itinerary segments, for historical data, named entity recognition technology from natural language processing and association rule algorithms from data mining are used to achieve automatic extraction and labeling of semantic tags;

[0129] Based on the basic tags of the future sightseeing vehicle itinerary segments, data in the database is retrieved. SQL join queries are executed in the database to retain data with the same basic tags, resulting in a preliminary filtered dataset. This is mainly to remove data and reduce workload.

[0130] For example, by searching for a road segment, the initial filtered dataset only retains data for that road segment, which can significantly reduce the amount of data for subsequent comparison and analysis.

[0131] To facilitate subsequent similarity calculations, the variables in the environment labels need to be one-hot encoded and converted into binary vectors. For example, sunny days are represented as [1,0,0,0,0,0,0] and cloudy days as [0,1,0,0,0,0,0]. This encoding method maps categorical variables to Euclidean space, which is convenient for subsequent similarity calculations.

[0132] Numerical labels in the environment label need to be normalized. The normalization process uses the min-max normalization method to map the data to the [0,1] interval. If the accuracy is to be improved, Z-score standardization can be used for further processing.

[0133] The environmental labels of the future sightseeing bus itinerary segment are compared with the environmental labels in the initially screened dataset, and the similarity is calculated using the following formula:

[0134]

[0135] In the formula: Simx represents the similarity score. It is an adjustable parameter. , where m is the number of dimensions for the environment tags. For the weight data of the i-th environmental label dimension, Let i be the value of the i-th environment label dimension in the environment label of the future itinerary segment. To initially filter the values ​​of the i-th environment label dimension in the historical itinerary sub-segments of the dataset, Let be the intermediate similarity, exp be an exponential function with the natural constant e as the base, and b be an adjustable parameter. .

[0136] The weighted processing method, which combines the weighted cosine similarity algorithm with the improved distance similarity, provides a more comprehensive reflection of the similarity between two environmental label sets. This improves the accuracy of the calculation and makes the results more representative.

[0137] When the value is close to 1, the final similarity depends more on the weighted cosine similarity; when... When the similarity approaches 0, the final similarity depends more on the improved Euclidean distance similarity.

[0138] The larger the b-value, the more significant the impact of distance on similarity; even if the numerical differences between two label sets are small, the similarity results will change significantly. The smaller the b-value, the smaller the impact of distance on similarity.

[0139] The weights 'b' are used to balance the weights of cosine similarity and distance similarity, while 'b' is used to adjust the degree of influence of distance on similarity.

[0140] When applying, When the value of b is high, it can be appropriately reduced to avoid excessive interference from distance in determining directional similarity. When the similarity is low, it is necessary to balance the reasonableness of distance scaling by b, so as to avoid the similarity result from approaching 0 or 1 too much.

[0141] Arrange the data in descending order of similarity, select the top K groups of similarity corresponding to the power generation, and then weight the power generation to obtain the power generation for each segment of the sightseeing bus trip. The weighting formula is as follows:

[0142]

[0143] In the formula, The predicted power generation for the sightseeing bus route segment. Let be the power generation of the j-th group, e be the natural constant, and c be the attenuation coefficient, where c > 0. The objective function for c is fitted using the least squares method. For the i-th The interval between the corresponding time point and the current time.

[0144] The principle behind this formula is that the closer the data is to the current time, the higher its reliability. This is because the longer the time is, the greater the changes in the external environment. Therefore, it is necessary to use data from different times to assign attenuation weights so that recent data contributes more to the prediction results. The weights decrease as the time gets closer to the present.

[0145] To highlight the importance of recent data, an exponential decay approach is adopted. For the decay factor, the least squares method is used for fitting. Regarding the least squares fitting, the computer processes the data, mainly transforming the objective function into a linear regression problem, minimizing the fitting error of logarithmic weights and time, and transforming it through the linear least squares formula to obtain the decay factor. Then, the decay factor can be directly substituted into the formula, so it will not be described in detail.

[0146] This invention describes a charging and discharging control system for the power battery of a new energy sightseeing vehicle. It works in conjunction with the existing sightseeing vehicle management system to collect travel data such as departure time, route, load, and driving style. It combines meteorological data to segment the journey and correlate weather data. After power generation prediction, demand analysis, and decision management, it generates a charging and discharging control strategy and feeds it back to the sightseeing vehicle management system. It uses an edge network to monitor the execution data in real time to form a closed loop, realizes dynamic matching of energy flow and operation scheduling, improves energy utilization efficiency and operational economy, has wide adaptability, good use effect, and has good application prospects.

[0147] After collecting relevant data and performing basic preprocessing and analysis, further demand analysis is needed to understand whether the device is discharging or charging. This demand analysis relies on the demand analysis module.

[0148] The demand analysis module constructs a random forest regression model, which processes future sightseeing bus trip data and future weather data and inputs them into the model to predict the energy consumption demand of the sightseeing bus in the future trip.

[0149] The steps for constructing a random forest regression model are as follows:

[0150] Acquire historical data, clean the collected data, and then encode the categorical data;

[0151] For example, check if there are missing or outlier values ​​in the data; fill missing values ​​with the mean, median, or by extrapolating from relevant data; and remove or correct outlier values.

[0152] For example, for categorical data such as driving routes, weather conditions (sunny, rainy, cloudy, etc.), and driver IDs, one-hot encoding or label encoding can be used to convert them into numerical data.

[0153] Historical data was processed using the min-max normalization method, and the processed data was divided into a training set and a test set with a ratio of 7:3.

[0154] From the training set, a training set for each decision tree is constructed by random sampling with replacement.

[0155] The decision tree nodes are constructed by randomly selecting a subset of features from all features. The optimal split point is found according to the criterion of minimizing the mean squared error. For example, if the number of features is 16, then when splitting, 4 features are randomly selected (one node is divided into two child nodes, and this process is repeated until the stopping condition is met, such as the number of node samples is less than a certain threshold or the depth of the tree reaches a preset value, etc.), and a decision tree is constructed to form a random forest regression model;

[0156] The random forest regression model is trained using a test set, and the mean squared error is used as the evaluation metric to output a random forest regression model whose test accuracy is greater than or equal to a preset threshold.

[0157] When making predictions, for each decision tree, the test sample is traversed from the root node down according to the node's splitting condition until a leaf node is reached. The value of the leaf node is the prediction value of that decision tree for the test sample. Then, the average of all prediction values ​​is calculated, and the average value is the output value of the random forest regression model.

[0158] For example: There exists a set of historical data for sightseeing vehicles, including driving routes, distances traveled, temperatures, driving ratings, number of passengers, and actual power consumption data.

[0159] First, the data is processed by performing one-hot encoding on the driving routes and normalizing other numerical data. Then, the data is divided into training and test sets. A random forest regression model is trained using the training set. Assuming that after hyperparameter tuning, the number of decision trees is determined to be 50 and the maximum depth is 10, the model is validated using the test set. If the mean squared error on the test set is 1.2 and the mean absolute error is 0.8, it indicates that the model has good predictive performance.

[0160] When a future sightseeing bus itinerary is "Route A, travel distance 20 kilometers", the future weather data is "temperature 27℃", the driver data is "driving style score 7 points", and the load data is "15 passengers", these data are input into the trained model, and the predicted power consumption output by the model is the predicted power demand of the sightseeing bus under this condition.

[0161] When understanding power generation and demand, comprehensive data analysis is required to formulate appropriate decisions, which depends on the decision management module.

[0162] Decision Management Module: Acquires power generation, energy consumption demand, and current power consumption data of sightseeing vehicles, calculates comprehensive control values, compares comprehensive control values ​​with corresponding control intervals, and formulates control strategies based on the comparison results;

[0163] Comprehensive control value = (Rated power of sightseeing vehicle + Power generation of all future sightseeing vehicle trip segments - Energy consumption demand - Safe power) / Rated power of sightseeing vehicle × 100%. The safe power is generally preset at 20% of the rated power of the sightseeing vehicle. The control range includes the charging range (-100%, BZ1) and the fluctuation range. And the discharge range (BZ2, 100%), the fluctuation range is as follows: .

[0164] The comprehensive control value is compared with the corresponding control range, and the control strategy is formulated based on the comparison results, including:

[0165] If the comprehensive control value is within the charging range, that is, the first charge adopts the nighttime off-peak charging, the sightseeing vehicle is charged to full charge, and the second charge adopts dynamic charging, then the control strategy is the first charge + dynamic charging.

[0166] This type of vehicle requires two charges as a single charge is insufficient to complete the trip. The arrangement of sightseeing vehicles is relatively reasonable during use. Under normal circumstances, the power demand of the sightseeing vehicles will not reach twice the rated power of the vehicles. Because there are a large number of sightseeing vehicles, they can be arranged reasonably.

[0167] Furthermore, the strategy proposed in this application for tourist sightseeing vehicles during off-peak hours in scenic areas will not consider discharging or electricity prices during peak hours; vehicles will be charged whenever possible to meet the operational needs of the scenic area.

[0168] If the comprehensive control value is not within the charging range, then the secondary control value is calculated. The secondary control value = (current power of the sightseeing vehicle + power generation of all future sightseeing vehicle trip segments - energy consumption demand - safe power) / rated power of the sightseeing vehicle × 100%;

[0169] This indicates that a full charge is sufficient for the vehicle's needs. However, since the vehicle is driven daily, it will not remain in a state of full charge. Further analysis is needed to determine whether the charge or discharge capacity is sufficient.

[0170] If the secondary control value is within the charging range, nighttime off-peak charging is adopted, and the charging amount = |secondary control value × sightseeing vehicle rated power|. At this time, the control strategy is single nighttime charging.

[0171] This indicates that the current battery level is low and needs to be replenished.

[0172] If the secondary control value is within the fluctuation range, the control strategy is not to perform charging and discharging treatment;

[0173] This situation indicates that the battery level is just right, so no action is taken.

[0174] If the secondary control value is within the discharge range, then discharge is performed. The discharge amount = secondary control value × sightseeing vehicle rated power, and dynamic discharge is executed. At this time, the control strategy is dynamic discharge.

[0175] This situation indicates that there is sufficient electricity. This situation usually occurs during the off-season for tourism combined with weekdays. From an economic point of view, selling the excess electricity back to the power grid maximizes economic benefits.

[0176] The dynamic charging and discharging scheme is optimized using a charging and discharging strategy model based on a deep Q-network.

[0177] The input to the charging and discharging strategy optimization model based on deep Q network during dynamic charging is the charging amount, dynamic electricity price data, and loss data corresponding to the charging speed. The output is the optimal charging power, charging time arrangement, and charging amount at each time.

[0178] The input to the dynamic discharge strategy optimization model based on deep Q-network is the discharge capacity, dynamic electricity price data, and loss data corresponding to the discharge rate. The output is the optimal discharge power, discharge time arrangement, and discharge amount at each time.

[0179] Dynamic charging and dynamic discharging use the same model architecture, only the input data and output results are different.

[0180] The model's decision logic layer learns the optimal charging strategy and the optimal discharging strategy based on the reward mechanism through continuous interaction between the agent and the environment.

[0181] The core of the model lies in planning to minimize the total charging cost and maximize the total discharging benefit.

[0182] Discharging supports V2G mode, feeding excess power back to the grid.

[0183] In other words, the model aims to minimize charging costs during dynamic charging and maximize discharge benefits during dynamic discharging. The model is subject to certain constraints, namely safe charging requirements, namely power range and temperature range.

[0184] By using optimization algorithms such as stochastic gradient descent, the policy parameters of the agent are continuously adjusted under the constraints to optimize the objective function, thereby determining the optimal charging and discharging power, time and energy at each time step.

[0185] This invention describes a charging and discharging control system for the power battery of a new energy sightseeing vehicle. During use, it improves energy management accuracy through multi-dimensional data integration and spatiotemporal modeling: it collects sightseeing vehicle trip data and meteorological data, uses time series analysis and spatial clustering algorithms to segment the trip into stationary / moving states, achieves deep coupling between trip segments and weather data through data fusion technology, and combines similarity retrieval and exponential weighting algorithms from a photovoltaic power generation prediction model to provide more granular spatiotemporal power generation prediction support with stronger environmental adaptability for charging and discharging decisions. This significantly improves the scientific nature and reliability of energy flow management for new energy sightseeing vehicles, resulting in good overall performance.

[0186] Once a corresponding charge and discharge control strategy is formulated, it will be strictly implemented. However, in actual use, it will be affected by external environmental and other human factors, which will cause certain changes in the implementation of the plan. Therefore, it is necessary to dynamically monitor the charge and discharge process, and this step relies on the dynamic monitoring module.

[0187] The dynamic monitoring module executes control strategies, acquires control strategy execution data in real time through the edge network, generates and applies optimization strategies through the dynamic monitoring model, and dynamically manages the charging and discharging of sightseeing vehicles.

[0188] Real-time acquisition of control strategy execution data via edge network: Sensors deployed on sightseeing vehicles and charging facilities using edge network are used to collect real-time data on charging and discharging temperatures, current electricity prices, charging power, and discharging power during the execution of charging and discharging strategies.

[0189] When data is collected and controlled through the edge network, it is transmitted to the edge server in real time. The edge server performs preliminary cleaning and preprocessing on the data, removes outliers and duplicate data, and stores the processed data in the local cache.

[0190] As a core component of the charging and discharging management of sightseeing vehicles in scenic areas, the dynamic monitoring model achieves real-time monitoring and dynamic optimization of the charging and discharging process through the collaborative operation of the data input layer, feature extraction layer, decision analysis layer, and strategy output layer. This model acquires data based on edge networks, combining spatiotemporal graph neural networks and deep Q-network technology, comprehensively considering both electricity price and charging temperature factors. Its aim is to reduce charging and discharging costs, improve energy utilization efficiency, and ensure the high efficiency and stability of sightseeing vehicle energy management.

[0191] The dynamic regulatory model includes:

[0192] Data input layer: Receives charging and discharging strategy execution data transmitted from the edge network and performs normalization processing;

[0193] For example, at a certain moment, the sensor collects data such as a charging power of 30kW, a remaining battery capacity of 40kWh, a charging temperature of 25℃, and a current electricity price of 0.8 yuan / kWh.

[0194] For example, if the charging power is normalized, with a historical minimum of 5kW and a maximum of 50kW, and the current charging power is 30kW, then the normalized charging power is 0.56. The normalized data will be used as the input to the feature extraction layer, providing a standardized data foundation for subsequent analysis.

[0195] Feature extraction layer: Utilizes a spatiotemporal graph neural network to extract spatiotemporal features from the data, and analyzes the correlation between charging and discharging states, electricity prices, and temperature in time series and spatial distribution;

[0196] Spatiotemporal graph neural networks are used to extract spatiotemporal features from the data. In charging and discharging management scenarios, the data not only exhibits time-series variation patterns but also spatial relationships between different sightseeing vehicles and charging facilities. Spatiotemporal graph neural networks model this through graph structures, treating each sightseeing vehicle or charging facility as a node in the graph. The connections between nodes represent their spatial relationships; for example, vehicles geographically close to each other may be affected by similar environmental factors. In the time dimension, recurrent neural networks process the time-series data of each node to capture the data's changing trends over time.

[0197] For example, by analyzing the correlation between charging and discharging states, electricity prices, and temperature in time series and spatial distribution, in terms of time series, spatiotemporal graph neural networks can learn the changing patterns of charging power at different times of the day, as well as the correlation between these changes and electricity price fluctuations and temperature changes.

[0198] For example, by studying historical data, it was found that when electricity prices are low and temperatures are suitable, vehicle charging power tends to be higher.

[0199] In terms of spatial distribution, spatiotemporal graph neural networks can analyze the differences in the charging and discharging states of vehicles in different regions, as well as the relationship between these differences and local electricity pricing policies and ambient temperature.

[0200] For example, in different parking lots within a scenic area, the charging and discharging behaviors of vehicles differ significantly due to variations in electricity prices and temperature distribution. By extracting and analyzing these spatiotemporal characteristics, more in-depth and valuable information can be provided to the decision-making analysis layer.

[0201] Decision Analysis Layer: Based on a deep Q-network decision mechanism, with the goal of minimizing charging and discharging costs and maximizing energy utilization efficiency, it combines the results of the feature extraction layer to calculate the Q value under different states, evaluate the merits of the current charging and discharging strategy, and adjust the strategy based on the dual factors of electricity price and charging temperature.

[0202] A two-factor strategy is adopted for adjustment: if the charging temperature is high, the charging speed is reduced; if the electricity price is low, the charging speed is reduced only slightly; and if the electricity price is high, the charging speed is reduced significantly.

[0203] Strategy output layer: Outputs optimization strategies, including discharge optimization strategies or charging optimization strategies;

[0204] Charging optimization strategies include charging power and charging time;

[0205] For example, it can output a power of 25kW for charging within the next 2 hours.

[0206] The discharge optimization strategy includes discharge power and discharge time.

[0207] For example, discharge at a power of 15kW within the next hour.

[0208] These optimization strategies will be transmitted to the sightseeing vehicle's charging and discharging management system via the edge network, guiding the vehicle's actual charging and discharging operations and achieving dynamic optimization management of the sightseeing vehicle's charging and discharging process. Simultaneously, the system will continuously monitor the effectiveness of the strategies and feed relevant data back to the data input layer, forming a closed-loop management system to continuously optimize charging and discharging strategies and improve management efficiency.

[0209] This invention describes a charging and discharging control system for the power battery of a new energy sightseeing vehicle. Through multi-source data fusion and a photovoltaic power generation prediction model, it achieves accurate prediction of power generation in spatiotemporal scenarios. Furthermore, it improves the accuracy of energy demand prediction by training cleaned and coded historical data using a random forest regression model. Based on comprehensive control values ​​and multi-interval thresholds combined with strategies such as off-peak electricity at night, dynamic charging and discharging, and V2G discharging, it achieves power balance and cost optimization. Further, it collects data in real time through an edge network, and generates optimization strategies using a spatiotemporal graph neural network and a deep Q-network, realizing multi-dimensional dynamic optimization of the charging and discharging of the new energy sightseeing vehicle. This significantly improves energy utilization efficiency, battery life, and grid coordination capabilities, demonstrating good performance and promising application prospects.

[0210] The weighting coefficients in the above formulas are determined using the coefficient of variation method. This method assigns weights to each indicator based on the degree of variation between its current value and the target value. If the numerical difference of an indicator is large, clearly distinguishing each evaluated object, it indicates that the indicator has rich discriminative information and should therefore be given a larger weight. Conversely, if the numerical difference of each evaluated object on a certain indicator is small, then the indicator's ability to distinguish each evaluated object is weak, and therefore it should be given a smaller weight. This method directly utilizes the information contained in each indicator to calculate the weight of the indicator, thus possessing objectivity.

[0211] In this application, the various formulas mentioned are all calculated by removing dimensions and taking their numerical values. The formulas are established by collecting a large amount of data and simulating the most recent real situation. Some coefficients or weights in the formulas are set by those skilled in the art according to the actual situation, so they will not be elaborated here.

[0212] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0213] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0214] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A new energy tourism sightseeing vehicle power battery charging and discharging control system, characterized in that: Comprise: Power generation prediction module: extract future trip and future weather data from sightseeing vehicle management system and meteorological platform, segment the future trip, and use data fusion technology to identify future weather data on future sightseeing vehicle trip sub-section, predict the power generation of all future sightseeing vehicle trip sub-sections through photovoltaic power generation prediction model; The steps of predicting the power generation of the sightseeing vehicle trip sub-section by the photovoltaic power generation prediction model are as follows: Obtain the semantic label of the future sightseeing vehicle trip sub-section, divide the semantic label into basic label and environment label, and the basic label is used to describe the basic attribute of the trip sub-section; Based on the basic label of the future sightseeing vehicle trip sub-section, search the database, retain the data with the same basic label, and get the preliminary screening data set; Compare the environment label of the future sightseeing vehicle trip sub-section with the environment label in the preliminary screening data set, calculate the similarity, and the specific formula is as follows: ； wherein: Simx is the similarity, is an adjustable parameter, m is the number of dimensions of the environmental label, is the weight data of the i-th environmental label dimension, is the value of the i-th environmental label dimension in the future trip sub-section environmental label, is the value of the i-th environmental label dimension in the historical trip sub-section environmental label in the preliminary screening data set, is the intermediate similarity, exp is the exponential function with the natural constant e as the base, and b is an adjustable parameter, ; Arrange according to the similarity from high to low, select the power generation corresponding to the first K groups of similarity, and perform weighted processing on the power generation to obtain the power generation of the sightseeing vehicle trip sub-section; Demand analysis module: build a random forest regression model, input the processed future sightseeing vehicle trip data and future weather data into the model, and predict the energy consumption demand of the future sightseeing vehicle trip; Decision management module: obtain the power generation, energy consumption demand and current power data of the sightseeing vehicle, calculate the comprehensive control value, compare the comprehensive control value with the corresponding control interval, and formulate the control strategy according to the comparison result; Dynamic supervision module: execute the control strategy, obtain the control strategy execution data in real time through the edge network, generate and apply the optimization strategy through the dynamic supervision model, and dynamically manage the charging and discharging of the sightseeing vehicle.

2. The power battery charge and discharge control system of a new energy tourism sightseeing vehicle according to claim 1, characterized in that: Standardized application program interface technology is used to collect future trip and future weather data. Future trip includes departure time, route data, estimated passenger load and driving style data. Future weather data includes temperature, humidity, wind speed and weather conditions.

3. The new energy tourism sightseeing vehicle power battery charge and discharge control system according to claim 2, characterized in that: Segmenting the future trip includes segmenting the sightseeing vehicle static state by combining time series analysis and segmenting the sightseeing vehicle driving state by using spatial clustering algorithm to obtain sightseeing vehicle trip sub-sections; The steps of using data fusion technology to identify future weather data on future sightseeing vehicle trip sub-sections are as follows: Deeply structure the future sightseeing vehicle trip sub-section data so that it contains unique identifier, start and end timestamp, spatial coordinate sequence, state type and associated road segment ID; Standardize the future weather data according to uniform time granularity; Based on the timestamp and spatial coordinates of the future sightseeing vehicle trip sub-section, associate with the future weather data; Based on spatial dimension interpolation and time dimension interpolation, process spatial data and time data; Use DTW algorithm to align the time sequence of the data caused by the time misalignment of the trip and the weather data due to the fluctuation of the vehicle driving speed; Build a semantic network, use natural language processing named entity recognition technology and data mining association rule algorithm to realize automatic extraction and labeling of semantic labels, and form the semantic labels of each future sightseeing vehicle trip sub-section.

4. The new energy tourism sightseeing vehicle power battery charge and discharge control system according to claim 3, characterized in that: The formula for weighting the power generation is as follows: ； In the formula, is the predicted power generation of the sightseeing vehicle trip sub-section, is the jth group power generation, e is a natural constant, c is a decay coefficient, c>0, and c is fitted using a least squares objective function, is the ith corresponds to the interval of the time point and the current time.

5. The new energy tourism sightseeing vehicle power battery charge and discharge control system according to claim 4, characterized in that: The steps of constructing the random forest regression model are as follows: Obtain historical data, clean the collected data, and then encode the category type data; Use the min-max normalization method to process the historical data, and divide the processed data into a training set and a test set; From the training set, construct the training set of each decision tree through random sampling with replacement; Randomly select a part of features from all features to construct the nodes of the decision tree, find the optimal split point according to the minimum mean square error criterion, construct the decision tree, and form the random forest regression model; Use the test set to train the random forest regression model, use the mean square error evaluation index, and output the random forest regression model with a test accuracy greater than or equal to a preset threshold.

6. The new energy tourism sightseeing vehicle power battery charge and discharge control system according to claim 5, characterized in that: The comprehensive regulation value = (the rated electric quantity of the sightseeing vehicle + the power generation of all future sightseeing vehicle trip sub-sections - the energy consumption demand - the safety electric quantity) / the rated electric quantity of the sightseeing vehicle x 100%, the regulation interval includes the charging interval (-100%, BZ1), the fluctuation interval and the discharging interval (BZ2, 100%).

7. The new energy tourism sightseeing vehicle power battery charge and discharge control system according to claim 6, characterized in that: Compare the comprehensive control value with the corresponding control interval, and develop a control strategy based on the comparison result, including: If the comprehensive control value is in the charging interval, the first charging is performed at night during the valley period, the sightseeing vehicle is charged to a full state, and the second charging is performed dynamically, and the control strategy is first charging + dynamic charging; If the comprehensive control value is not in the charging interval, calculate the secondary control value, secondary control value = (current sightseeing vehicle power + all future sightseeing vehicle travel sub-section power generation - energy consumption demand - safety power) / rated sightseeing vehicle power * 100%; If the secondary control value is in the charging interval, use the night valley charging, and the charging power = |secondary control value * rated sightseeing vehicle power|, and the control strategy is single night charging; If the secondary control value is in the fluctuation interval, the control strategy is not charging and discharging; If the secondary control value is in the discharging interval, discharge the power = secondary control value * rated sightseeing vehicle power, and perform dynamic discharging, and the control strategy is dynamic discharging.

8. The new energy tourism sightseeing vehicle power battery charge and discharge control system according to claim 7, characterized in that: The dynamic charging and dynamic discharging schemes are based on a deep Q network-based charging and discharging strategy optimization model; The input of the deep Q network-based charging and discharging strategy optimization model during dynamic charging is charging power, dynamic electricity price data, and charging speed corresponding loss data, and the output is the optimal charging power, charging time arrangement, and charging power at each time; The input of the deep Q network-based charging and discharging strategy optimization model during dynamic discharging is discharging power, dynamic electricity price data, and discharging speed corresponding loss data, and the output is the optimal discharging power, discharging time arrangement, and discharging power at each time.

9. The new energy tourism sightseeing vehicle power battery charge and discharge control system according to claim 8, characterized in that: Real-time acquisition of control strategy execution data through edge network is to use sensors deployed on sightseeing vehicles and charging facilities through edge network to collect charging temperature, discharging temperature, current electricity price, charging power, and discharging power during charging and discharging strategy execution in real time; When collecting control strategy execution data through edge network, the data is transmitted to the edge server in real time through edge network, the edge server performs preliminary cleaning and preprocessing on the data, removes abnormal values and duplicate data, and stores the processed data in the local cache.

10. The new energy tourism sightseeing vehicle power battery charge and discharge control system according to claim 9, characterized in that: The dynamic supervision model includes: Data input layer: receives the charging and discharging strategy execution data transmitted by the edge network, and performs normalization processing; Feature extraction layer: use spatio-temporal graph neural network to extract the spatio-temporal features in the data, analyze the correlation between charging and discharging state, electricity price, and temperature in time series and spatial distribution; The decision analysis layer is a decision mechanism based on a deep Q network, aims to minimize the charging and discharging cost and maximize the energy utilization efficiency, combines the result of the feature extraction layer, calculates the Q value in different states, evaluates the advantages and disadvantages of the current charging and discharging strategy, and adjusts the strategy according to the electricity price and the charging temperature double factors. The strategy output layer outputs the optimized strategy, and the optimized strategy includes a discharging optimization strategy or a charging optimization strategy. The charging optimization strategy includes charging power and charging time. The discharging optimization strategy includes discharging power and discharging time.

Images